Sparsezoo stdout fix

[rahul-tuli]

This PR reverts commit https://github.com/neuralmagic/sparsezoo/commit/960cae23e8273cc7e3cfff9ceb65a8f3ecb8628f.

The newly added GoogleAnalytics class was throwing 429 Client Errors: a proposed fix for that was to suppress sys.stdout and sys.stderr streams in a multithreaded function, The problem with this fix was that the streams were not being restored properly and were suppressing output from all commands, and raised a 120 error code a potential fix could be to

• Execute this function in a single thread of execution, in which case the streams would be restored properly but we would lose the benefits of multithreading ❌
• Changing code to include a flush method in the NullDevice class, this would return an error code of zero but would still suppress all output ❌
• The proposed fix should have never made it in as the code was already handling all exceptions raised by the request ✅ + the fix was solving a plumbing problem rather than addressing the original issue.

I wasn't able to reproduce the exact error 429 Client Error from GoogleAnalytics request but raised a dummy HTTPError myself and verfied the the code actually catches all HTTPErrors, on another note this PR also makes the exception catching more specific over catching all Exceptions

If the 429 Client Errors re-appear they should be handled by the current code; otherwise should be addressed differently preferrably in the _send_request function without redirection of stdout/stderr streams.

$ deepsparse.check_hardware              
GenuineIntel CPU detected with 10 cores. (1 sockets with 10 cores each)
DeepSparse FP32 model performance supported: True.
DeepSparse INT8 (quantized) model performance supported: TRUE (emulated).

Non VNNI system detected. Performance speedups for INT8 (quantized) models is available, but will be slower compared with a VNNI system. Set NM_FAST_VNNI_EMULATION=True in the environment to enable faster emulated inference which may have a minor effect on accuracy.

Additional CPU info: {'L1_data_cache_size': 32768, 'L1_instruction_cache_size': 32768, 'L2_cache_size': 1048576, 'L3_cache_size': 14417920, 'architecture': 'x86_64', 'available_cores_per_socket': 10, 'available_num_cores': 10, 'available_num_hw_threads': 20, 'available_num_numa': 1, 'available_num_sockets': 1, 'available_sockets': 1, 'available_threads_per_core': 2, 'bf16': False, 'cores_per_socket': 10, 'dotprod': False, 'i8mm': False, 'isa': 'avx512', 'num_cores': 10, 'num_hw_threads': 20, 'num_numa': 1, 'num_sockets': 1, 'threads_per_core': 2, 'vbmi': False, 'vbmi2': False, 'vendor': 'GenuineIntel', 'vendor_id': 'Intel', 'vendor_model': 'Intel(R) Core(TM) i9-7900X CPU @ 3.30GHz', 'vnni': False, 'zen1': False}

---

• To see the specific tasks where the Asana app for GitHub is being used, see below:
  • https://app.asana.com/0/0/1204714095366468
  • https://app.asana.com/0/0/1204698591832759
  • https://app.asana.com/0/0/1204680191909106